Elevated blood urea nitrogen is associated with recurrence of post-operative chronic subdural hematoma

Synthetic minority over-sampling technique-enhanced machine learning models for predicting recurrence of postoperative chronic subdural hematoma

Objective

Chronic subdural hematoma (CSDH) is a neurological condition with high recurrence rates, primarily observed in the elderly population. Although several risk factors have been identified, predicting CSDH recurrence remains a challenge. Given the potential of machine learning (ML) to extract meaningful insights from complex data sets, our study aims to develop and validate ML models capable of accurately predicting postoperative CSDH recurrence.

Methods

Data from 447 CSDH patients treated with consecutive burr-hole irrigations at Wenzhou Medical University's First Affiliated Hospital (December 2014-April 2019) were studied. 312 patients formed the development cohort, while 135 comprised the test cohort. The Least Absolute Shrinkage and Selection Operator (LASSO) method was employed to select crucial features associated with recurrence. Eight machine learning algorithms were used to construct prediction models for hematoma recurrence, using demographic, laboratory, and radiological features. The Border-line Synthetic Minority Over-sampling Technique (SMOTE) was applied to address data imbalance, and Shapley Additive Explanation (SHAP) analysis was utilized to improve model visualization and interpretability. Model performance was assessed using metrics such as AUROC, sensitivity, specificity, F1 score, calibration plots, and decision curve analysis (DCA).

Results

Our optimized ML models exhibited prediction accuracies ranging from 61.0% to 86.2% for hematoma recurrence in the validation set. Notably, the Random Forest (RF) model surpassed other algorithms, achieving an accuracy of 86.2%. SHAP analysis confirmed these results, highlighting key clinical predictors for CSDH recurrence risk, including age, alanine aminotransferase level, fibrinogen level, thrombin time, and maximum hematoma diameter. The RF model yielded an accuracy of 92.6% with an AUC value of 0.834 in the test dataset.

Conclusion

Our findings underscore the efficacy of machine learning algorithms, notably the integration of the RF model with SMOTE, in forecasting the recurrence of postoperative chronic subdural hematoma. Leveraging the RF model, we devised an online calculator that may serve as a pivotal instrument in tailoring therapeutic strategies and implementing timely preventive interventions for high-risk patients.

Biofluid Biomarkers in the Prognosis of Chronic Subdural Hematoma: A Systematic Scoping Review

The present systematic scoping review aimed at mapping and analyzing the available literature on biological fluid (biofluid) biomarkers showing promise in the prediction of chronic subdural hematoma (cSDH) recurrence and the prognosis of neurological/functional patient outcome. Twenty-three studies published between 2003 and 2023 investigating a diverse range of biomarkers in hematoma fluid and/or the circulation in 3749 patients were included. Immune cell populations and inflammatory/anti-inflammatory cytokines comprised the most studied category of biomarkers displaying significant findings. A notable time trend in biomarker studies was a recent shift in research focus towards the analysis of circulating biomarkers. Several biomarkers were indicated as independent predictors of cSDH recurrence and/or functional/neurological outcome, including circulating fibrinogen degradation products (FDP), brain natriuretic peptide (BNP-1) and high-density lipoprotein (HDL), as well as blood urea nitrogen (BUN) and the ratios of blood neutrophil to lymphocyte (NLR) or red blood cell distribution width to platelet count (RPR). While studies on cSDH prognostic biomarkers have gained, in recent years, momentum, additional multicenter prospective studies are warranted to confirm and extend their findings. The identification of prognostic biofluid biomarkers in cSDH is an active field of research that may provide future tools, guiding clinical decisions and allowing for the design of treatments based on risk stratification.

A low prognostic nutritional index is associated with chronic subdural hematoma recurrence

Chronic subdural hematoma (CSDH) is a common neurosurgical condition. Inflammation has been found to play an essential role in the formation of CSDHs, and the prognostic nutritional index (PNI), a nutritional and inflammatory baseline marker, plays a role in predicting the prognosis of many diseases. We aimed to identify the relationship between PNI and CSDH recurrence. This study retrospectively analyzed 261 CSDH patients who underwent burr hole evacuation in Beijing Tiantan Hospital from August 2013 to March 2018. The PNI was calculated as 5 ∗ lymphocyte count (10⁹/L) + serum albumin concentration (g/L), and these markers were obtained from the peripheral blood test on the day of discharge from the hospital. Recurrence was defined as operated hematoma enlargement accompanied by newly emerging neurological disorders. The comparison of baseline characteristics demonstrated that patients with bilateral hematoma and low levels of albumin, lymphocytes, and PNI were more likely to be recurrent. After adjusting for age, sex, and other important variables, decreased PNI levels were associated with an increased risk of CSDH (OR, 0.803, 95% CI: 0.715-0.902, p = 0.001). The addition of PNI to conventional risk factors significantly improved the risk prediction of CSDH (net reclassification index: 71.12%, p = 0.001; integrated discrimination index: 10.94%, p = 0.006). A low PNI level is associated with an increased risk of CSDH recurrence. As an easily obtainable nutritional and inflammatory marker, PNI may play a significant role in predicting the recurrence of CSDH patients.

Nutritional and inflammatory peripheral blood markers for risk assessment of chronic subdural hematoma: a case-control study

BACKGROUND

Some peripheral blood markers have been demonstrated to be correlated with the re-formation of chronic subdural hematoma (CSDH). The aim of this study was to identify the correlation between nutritional/inflammatory peripheral blood markers and CSDH.

METHODS

188 CSDH patients and 188 age-matched healthy controls were included in this research. The clinical characteristics and peripheral blood markers associated with nutritional or inflammatory status were obtained and analyzed. Conditional logistic regression analysis was applied to identify the potential CSDH risk factors. All the participants were divided into 3 groups based on the tertiles of change in risk factors. The Cochran-Armitage test and one way ANOVA were applied to identify the association between baseline characteristics and independent risk factors. Moreover, the net reclassification index (NRI) and integrated discrimination index (IDI) were calculated to evaluate the improvement in model performance after adding the independent risk factors in the conventional model.

RESULTS

The logistic regression analysis demonstrated that the increased albumin (OR, 0.615; 95 %CI,0.489-0.773; P < 0.001) and lymphocyte count (OR, 0.141; 95 %CI,0.025-0.796; P = 0.027) were associated with lower risk of CSDH. Moreover, addition of albumin and lymphocyte to conventional risk factors significantly improved the risk prediction of CSDH(NRI: 46.47 %, P < 0.001; IDI: 30.92 %, P < 0.001; NRI: 22.45 %, P = 0.027; IDI: 1.23 %, P = 0.037, respectively) CONCLUSION: The decreased albumin and lymphocyte levels were correlated with a high risk of chronic subdural hematoma. The nutritional and inflammatory serum markers should be put great attention because these markers may play roles in finding the cause of CSDH and predicting its risk.

Non-invasive Liver Fibrosis Scores Are Associated With Recurrence of Postoperative Chronic Subdural Hematoma

Objective

Although liver diseases have already been identified as a risk factor for increased recurrence and mortality in patients with chronic subdural hematoma (CSDH), the association between subclinical liver disease, specifically liver fibrosis (LF), and CSDH remains unknown. In the present study, we aimed to investigate the association between the LF scores and CSDH recurrence.

Methods

We retrospectively analyzed consecutive patients with CSDH who underwent burr-hole irrigation in the First Affiliated Hospital of Wenzhou Medical University between January 2015 and December 2018. The clinical data were collected, and the LF scores were calculated including aspartate aminotransferase–platelet ratio index (APRI), fibrosis-4 (FIB-4), and Forns index. Multivariable logistic regression analysis was applied to identify the association between the LF scores and CSDH recurrence, and Cox regression model and Fine–Gray competing risks model were performed to calculate hazard ratios (HRs) for CSDH recurrence based on time-to-event outcomes. The C-statistic, the integrated discrimination improvement (IDI), and the net reclassification improvement (NRI) evaluated the additive value of the LF scores to predict the recurrence of CSDH.

Results

A total of 419 patients with CSDH were included, hematoma recurrence was observed in 62 patients (14.80%) within 1 year after surgery. The LF scores were significantly higher in those who recurred, whereas the standard hepatic assays were mostly normal. The patients were assigned to groups of high and low LF scores based on the validated cut-offs; compared with the subjects with low scores, those with high score levels had significantly higher recurrence rates. After adjusting for potential confounders, the LF scores were independently associated with CSDH recurrence, multivariable-adjusted HRs (95% CI) for those with higher levels of APRI, FIB-4, and Forns score were 4.32 (1.37–13.60), 2.56 (1.20–5.43), and 2.02 (1.07–3.79) for the recurrence of CSDH, respectively. Moreover, adding the APRI to the conventional model improved the C-statistic from 0.731 to 0.763, with an NRI and IDI of 7.50 and 1.35%, respectively. Two further commonly-used LF score indices (FIB-4 score and Forns index) yielded comparable results.

Conclusions

The data from this study first indicated that the high LF scores were significantly associated with the recurrence of CSDH and that careful follow-up in these patients may be needed.

Relevance of comorbidities and antithrombotic medication as risk factors for reoperation in patients with chronic subdural hematoma

In an aging Western society, the incidence of chronic subdural hematomas (cSDH) is continuously increasing. In this study, we reviewed our clinical management of cSDH patients and identified predictive factors for the need of reoperation due to residual or recurrent hematomas with a focus on the use of antithrombotic drugs. In total, 623 patients who were treated for cSDH with surgical evacuation between 2006 and 2016 at our department were retrospectively analyzed. Clinical and radiological characteristics and laboratory parameters were investigated as possible predictors of reoperation with univariate and multivariate analyses. Additionally, clinical outcome measures were compared between patients on anticoagulants, on antiplatelets, and without antithrombotic medication. In univariate analyses, patients on anticoagulants and antiplatelets presented significantly more often with comorbidities, were significantly older, and their risk for perioperative complications was significantly increased. Nevertheless, their clinical outcome was comparable to that of patients without antithrombotics. In multivariate analysis, only the presence of comorbidities, but not antithrombotics, was an independent predictor for the need for reoperations. Patients on antithrombotics do not seem to necessarily have a significantly increased risk for residual hematomas or rebleeding requiring reoperation after cSDH evacuation. More precisely, the presence of predisposing comorbidities might be a key independent risk factor for reoperation. Importantly, the clinical outcomes after surgical evacuation of cSDH are comparable between patients on anticoagulants, antiplatelets, and without antithrombotics.